Auto merge of #143182 - xdoardo:more-addrspace, r=workingjubilee

Allow custom default address spaces and parse `p-` specifications in the datalayout string

Some targets, such as CHERI, use as default an address space different from the "normal" default address space `0` (in the case of CHERI, [200 is used](https://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-877.pdf)). Currently, `rustc` does not allow to specify custom address spaces and does not take into consideration [`p-` specifications in the datalayout string](https://llvm.org/docs/LangRef.html#langref-datalayout).

This patch tries to mitigate these problems by allowing targets to define a custom default address space (while keeping the default value to address space `0`) and adding the code to parse the `p-` specifications in `rustc_abi`. The main changes are that `TargetDataLayout` now uses functions to refer to pointer-related informations, instead of having specific fields for the size and alignment of pointers in the default address space; furthermore, the two `pointer_size` and `pointer_align` fields in `TargetDataLayout` are replaced with an `FxHashMap` that holds info for all the possible address spaces, as parsed by the `p-` specifications.

The potential performance drawbacks of not having ad-hoc fields for the default address space will be tested in this PR's CI run.

r? workingjubilee

crates/hir-ty/src/layout.rs

@@ -261,7 +261,7 @@ pub fn layout_of_ty_query(
         }
         // Potentially-wide pointers.
         TyKind::Ref(_, _, pointee) | TyKind::Raw(_, pointee) => {
-            let mut data_ptr = scalar_unit(dl, Primitive::Pointer(AddressSpace::DATA));
+            let mut data_ptr = scalar_unit(dl, Primitive::Pointer(AddressSpace::ZERO));
             if matches!(ty.kind(Interner), TyKind::Ref(..)) {
                 data_ptr.valid_range_mut().start = 1;
             }
@@ -285,7 +285,7 @@ pub fn layout_of_ty_query(
                     scalar_unit(dl, Primitive::Int(dl.ptr_sized_integer(), false))
                 }
                 TyKind::Dyn(..) => {
-                    let mut vtable = scalar_unit(dl, Primitive::Pointer(AddressSpace::DATA));
+                    let mut vtable = scalar_unit(dl, Primitive::Pointer(AddressSpace::ZERO));
                     vtable.valid_range_mut().start = 1;
                     vtable
                 }

crates/hir-ty/src/layout/target.rs

@@ -2,7 +2,7 @@
 
 use base_db::Crate;
 use hir_def::layout::TargetDataLayout;
-use rustc_abi::{AlignFromBytesError, TargetDataLayoutErrors};
+use rustc_abi::{AlignFromBytesError, TargetDataLayoutErrors, AddressSpace};
 use triomphe::Arc;
 
 use crate::db::HirDatabase;
@@ -12,7 +12,7 @@ pub fn target_data_layout_query(
     krate: Crate,
 ) -> Result<Arc<TargetDataLayout>, Arc<str>> {
     match &krate.workspace_data(db).data_layout {
-        Ok(it) => match TargetDataLayout::parse_from_llvm_datalayout_string(it) {
+        Ok(it) => match TargetDataLayout::parse_from_llvm_datalayout_string(it, AddressSpace::ZERO) {
             Ok(it) => Ok(Arc::new(it)),
             Err(e) => {
                 Err(match e {
@@ -39,6 +39,7 @@ pub fn target_data_layout_query(
                         target,
                     } => format!(r#"inconsistent target specification: "data-layout" claims pointers are {pointer_size}-bit, while "target-pointer-width" is `{target}`"#),
                     TargetDataLayoutErrors::InvalidBitsSize { err } => err,
+                    TargetDataLayoutErrors::UnknownPointerSpecification { err } => format!(r#"use of unknown pointer specifer in "data-layout": {err}"#),
                 }.into())
             }
         },

crates/hir-ty/src/mir/eval.rs

@@ -630,7 +630,7 @@ impl Evaluator<'_> {
             Ok(target_data_layout) => target_data_layout,
             Err(e) => return Err(MirEvalError::TargetDataLayoutNotAvailable(e)),
         };
-        let cached_ptr_size = target_data_layout.pointer_size.bytes_usize();
+        let cached_ptr_size = target_data_layout.pointer_size().bytes_usize();
         Ok(Evaluator {
             target_data_layout,
             stack: vec![0],